In model-based image data segmentation, results of the segmentation comprise a model adapted to the image data. Such a model typically includes many details. Therefore, when viewing the model adapted to the image data, said model cannot be easily analyzed on a display or in a report. Adapting a simplified model representing and possibly emphasizing an aspect of a structure of interest to the image data is not feasible because a simplified model does not include enough detail to be successfully adapted to the image data. On the other hand, the simplified model which is not adapted to the structure of interest does not include the structure-specific information.